HKNAS: Classification of Hyperspectral Imagery Based on Hyper Kernel Neural Architecture Search (TNNLS 2023)
Pytorch implementation of our paper for Neural Architecture Search based hyperspectral image classification.
Fig.1 - Search Space. |
Fig.2 - Search Algorithm. |
- Install Pytorch 1.9 with Python 3.8.
- Clone this repo.
git clone https://github.com/DotWang/HKNAS.git
- For 3-D HK-CLS and 3-D HK-SEG, setting the 3-D convolution form in main.py
- Search, Training, Validation, Testing and Predicion (Taka an example of WHU-Hi-HongHu dataset):
- 1-D HK-CLS
cd 1DHKCLS
CUDA_VISIBLE_DEVICES=0 python main.py --flag 'honghu' --exp_num 10 --block_num 3 --layer_num 1
- 3-D HK-CLS
cd 3DHKCLS
CUDA_VISIBLE_DEVICES=0 python main.py --flag 'honghu' --exp_num 10 --block_num 3 --layer_num 3
- 3-D HK-SEG
cd 3DHKSEG
CUDA_VISIBLE_DEVICES=0 python main.py --flag 'honghu' --exp_num 10 --block_num 3 --layer_num 1
@ARTICLE{hknas,
author={Wang, Di and Du, Bo and Zhang, Liangpei and Tao, Dacheng},
journal={IEEE Transactions on Neural Networks and Learning Systems},
title={HKNAS: Classification of Hyperspectral Imagery Based on Hyper Kernel Neural Architecture Search},
year={2023},
volume={},
number={},
pages={1-15},
doi={10.1109/TNNLS.2023.3270369}}
[1] Pixel and Patch-level Hyperspectral Image Classification
Adaptive Spectral–Spatial Multiscale Contextual Feature Extraction for Hyperspectral Image Classification, IEEE TGRS, 2020 | Paper | Github
Di Wang∗, Bo Du, Liangpei Zhang and Yonghao Xu
[2] Image-level/Patch-free Hyperspectral Image Classification
Fully Contextual Network for Hyperspectral Scene Parsing, IEEE TGRS, 2021 | Paper | Github
Di Wang∗, Bo Du, and Liangpei Zhang
[3] Graph Convolution based Hyperspectral Image Classification
Spectral-Spatial Global Graph Reasoning for Hyperspectral Image Classification, IEEE TNNLS, 2023 | Paper | Github
Di Wang∗, Bo Du, and Liangpei Zhang
[4] ImageNet Pretraining and Transformer based Hyperspectral Image Classification
DCN-T: Dual Context Network with Transformer for Hyperspectral Image Classification, IEEE TIP, 2023 | Paper | Github
Di Wang∗, Jing Zhang, Bo Du, Liangpei Zhang, and Dacheng Tao